Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Virtually all penal laws in the world include a ­prohibition against homicide. Homicide as a general term refers to all cases in which a human being kills another human being by his or her own act, omission, or procurement. However, not all homicides carry criminal culpability. Sometimes, a killing may be committed in self-defense or by accident or any other circumstances that the law permits. These are excusable homicides or justifiable homicides. In this chapter, only criminal homicide is discussed and analyzed.

Countries differ considerably in how they define homicide and its categories. Some jurisdictions further divide criminal homicide into categories such as murder and manslaughter. There are also categories of homicide based on the level of intent, such as intentional homicide and negligent homicide. Acts such as infanticide, abortion, assisted suicide, and euthanasia may be either included or excluded from the national legal definition. Therefore, caution is usually needed in comparing and explaining homicide data across countries.

This chapter first introduces the data sources of international homicide data and then describes the recent homicide trends in Asian countries. Based on empirical research, this chapter also explores some correlates of homicide trends. In addition, homicide clearance is also discussed within a comparative approach.

2 International Homicide Data

Homicide data are usually considered the most valid and reliable data for cross-national comparison among all internationally available violence data (Marshall and Block 2004). Comparative research on homicide has often used homicide data from three international sources, including the International Criminal Police Organization (Interpol), the United Nations Surveys of Crime Trends and Operations of Criminal Justice Systems (UN-CTS), and the World Health Organization (WHO). The first two sources are cross-national criminal justice data sources, and the last one is a cross-national public health data source.

These three data sources often do not provide similar homicide estimates. This is partly due to definitional differences (Marshall and Block 2004). According to Marshall and Block (2004), Interpol uses the term murder rather than homicide. Murder refers to any act performed with the purpose of taking human life. This definition excludes manslaughter and abortion but not infanticide. However, after 1978 manslaughter is not explicitly excluded from the definition. In addition, Interpol does not provide data on ­unintentional homicide. In contrast, the UN-CTS survey includes a few categories such as total homicides, total intended homicides, attempted intended homicides, and unintentional homicides. Different countries may include or exclude euthanasia, infanticide, assisted suicide, or assault resulting in death. Unlike Interpol and UN data which collect homicide data from criminal justice agencies in member nations, WHO data are based on body counts (i.e., on cause-of-death reports submitted by participating nations) instead of incidents. Further, WHO homicide data do not include attempted or unintentional homicides (Marshall and Block 2004). Among these three sources of cross-national comparative homicide data, the WHO data are widely believed to be the most valid (Gartner 1990; Kalish 1988; Messner and Rosenfeld 1997).

The United Nations Office on Drugs and Crime (UNODC) has developed the International Homicide Statistics (IHS) which provides homicide statistics in 198 countries and territories from 2003 to 2008 (Harrendorf et al. 2010). The most recent UNODC homicide statistics can be found in its new publication, The 2011 Global Study on Homicide. The objective of this international statistics is to improve the availability of crime and criminal justice information and to expand the knowledge of crime trends. UNODC collects homicide data from a variety of data sources, including data collected by UN-CTS, data collected by other cross-national crime statistics sources such as Interpol and Statistical Office of the European Communities, international public health data such as data collected by WHO and Pan American Health Organization, and 33 national law enforcement and criminal justice data sources. One of the strengths of this dataset is that all sources used are fully referenced in the dataset and a methodological description is also provided (Harrendorf et al. 2010).

According to the IHS dataset, in 2004 the world average homicide rate was 7.6 per 100,000 population. Homicide rates (per 100,000 population) in 2004 for the 16 subregions of East Africa, North Africa, South America, Central Asia and Transcaucasian Countries, East Asia, Near and Middle East/South West Asia, South Asia, East Europe, South East Europe, West and Central Europe, and Oceania are presented in the map in Fig. 2.1. The map also suggests that the highest homicide levels are found in the Americas and Africa region, with the lowest homicide levels generally in countries of Europe. Homicide rates in Asia are also relatively low. The homicide data by country shows that in 2004 Kazakhstan, Cambodia, Democratic People’s Republic of Korea, Myanmar, and Philippines had the highest estimated homicide rates in Asia, while Hong Kong, Japan, Qatar, and United Arab Emirates had the lowest.

Fig. 2.1
figure 00021

Intentional homicide, rate per 100,000 population, by subregion, 2004 (retrieved online from http://www.unodc.org/images/data-and-analysis/homicide_rate_map.pdf)

Full size image

It should be noted that the IHS dataset does not include killings in armed conflict committed by groups of up to several hundred members. Therefore, the homicide data presented here should be interpreted with caution in countries affected by armed conflict. It is also important to note that the IHS dataset is about intentional homicide, that is, unlawful death purposefully inflicted on a person by another person. In other words, negligent homicides are not included. Thus, the method used by IHS captures the most basic commonality between the various data sources. Further, because of differences in the definition of homicide, countries may or may not count offenses such as assault leading to death, euthanasia, infanticide, or assistance with suicide in the homicide data. As a result, the differences in intentional homicide data here between countries and regions reflect not only the different numbers of killings across areas but also the extent to which countries and regions classify different types of killing as homicide (UNODC 2004).

3 Recent Homicide Trends in Asia

The UNODC’s (2010) international homicide data can be used to depict the changes in homicide rates over time in many countries. However, as explained above, comparisons between countries should be made with caution. In addition, data recording method in one country may also change over time. The average intentional homicide rates for countries in Asia and Oceania from 2003 to 2008 are presented in Fig. 2.2. Due to the incomplete information of some countries, not all the countries are included. The homicide trends in recent years suggest that on average, homicide rates are going down slowly, and there have not been abrupt increases or decreases from year to year. The country level data collected by UNODC are also grouped by subregion and presented in Tables 2.1– 2.4. Because of the discrepancies in the police data and public health data, both data are presented when available. Information of the data sources for each country can be obtained at the website of UNODC. Homicide rate is the number of homicide per 100,000 population.

Fig. 2.2
figure 00022

Average international homicide rates for countries in Asia and Oceania (2003–2008) (Harrendorf et al. 2010)

Full size image
Table 2.1 Homicide trends in Central Asia and Transcaucasian countries, 2003–2008 (UNODC 2010)
Full size table
Table 2.2 Homicide trends in East Asia, 2003–2008 (UNODC 2010)
Full size table
Table 2.3 Homicide trends in near Middle East/South West Asia, 2003–2008 (UNODC 2010)
Full size table
Table 2.4 Homicide trends in South Asia, 2003–2008 (UNODC 2010)
Full size table

4 Social Factors and Homicide

The macro-level empirical research on homicide has described and analyzed a number of factors that are correlated with homicide rates, and majority of this body of research has studied the homicide rates over time with a comparative approach. Some Asian countries and regions such as Japan, Singapore, Hong Kong, and Taiwan are often selected in the comparative research, and the social factors that have been studied in the literature include availability of firearms, use of capital punishment, youth population, population diversity, democracy, modernization, social stress and support, and many others. This section reviews the empirical literature that has studied homicide rates in Asian countries.

5 Use of Firearm in Homicide

The availability of firearms often plays an important role in homicide rates. In the UNODC 2010 report of International Statistics of Crime and Justice, it is found that areas with high homicide rates are usually the areas with the highest percentage homicide with firearms (Harrendorf et al. 2010). Data from Fig. 2.3 suggest that Asian countries are similar to European countries in terms of the percentage of homicides with firearm. In contrast, Americas have the highest percentage of homicides with firearm in the world.

Fig. 2.3
figure 00023

Percent of homicide committed with a firearm, 2003–2008 (Harrendorf et al. 2010)

Full size image

In a cross-national study on homicide among children, Krug et al. (1998) specifically examined the homicide rates among children in the USA with those in other countries. In this study, the selected countries were high-income countries defined by World Bank in 1994, among which Asian countries and regions included Singapore, Japan, Israel, Taiwan, and Hong Kong. The data about childhood homicide were between 1990 and 1995. Krug et al. found that among the Asian countries selected, Israel had the highest firearm-related homicide rates for children, while other Asian countries in the study had the lowest firearm-related homicide rates. However, inconsistent with the findings reported by Harrendorf et al. (2010), Israel did not have the highest total homicide rate. In fact, among the Asian countries studied, Singapore had the highest nonfirearm-related homicide rate for children and also the highest total homicide rate for children.

6 Capital Punishment and Homicide

Research using homicide data has explored the deterrent effect of capital punishment on homicide. The major concern of this body research is whether the threat of execution deters homicide better than other criminal sanctions. As the most serious violent crime, homicide usually receives the harshest punishment in the criminal justice system, and when available, death penalty is often used. Often, the public are strongly supportive of a death penalty for murder (Johnson 2006).

According to Zimring and Johnson (2008), Asian countries provide a better context to explore the deterrent effect of capital punishment. First, there is a great variety of political systems and policies toward capital punishment in Asia. Second, many Asian countries are undergoing rapid changes in death penalty laws and practices. Third, Asian countries are less influenced by the European culture, which provides more social contexts for comparative analyses and theory generalization. Fourth, most of the executions in the world are carried out in Asia, ranging from 85 to 95% in recent years. In addition, about 95% of Asian populations live in the jurisdictions which use capital punishment (Zimring and Johnson 2008).

Zimring et al. (2010) studied the deterrent effects of capital punishment on homicide trends in Asia. Specifically, this study used Hong Kong as a comparison city to examine the impacts of executions on homicide rates in Singapore over time. According to Zimring et al., there are similarities between Singapore and Hong Kong with regard to many aspects including population density, annual population growth, economic growth, adult literacy, birth rate, migrants-population ratio, and life expectancy. In addition, homicide levels and trends are remarkably similar in these two cities over 35 years after 1973.

In Singapore, offenders convicted of murder will be sentenced to death. Zimring et al. (2010) combined different sources of data and described that murder executions in Singapore rose in 1992 and dropped back quickly after the 1994–1995 peak. Further, homicide trials in Singapore often take a few months with a speedy appeal process after conviction. In contrast, Hong Kong had no executions from 1967 to 1998, and since 1993, mandatory life sentences are imposed for murder.

To explore the relationship between execution rates and homicide rates in Singapore, Zimring et al. (2010) used the murder rate in Hong Kong as a control variable in the statistical analyses. They estimated 14 models of homicide trend in Singapore during 1973–2007 and its shorter trend during 1981–2007. They concluded that the changes in execution levels had no significant impacts on the homicide trend in Singapore. Specifically, the decline in homicide rate happened before the sudden increase in executions in 1992, and the changes of homicide rate over time were largely unrelated to the extent of execution.

7 Youth Population and Homicide

Youth violence is a widespread problem around the world, and the highest homicide rate has been usually found among males between the ages of 15 and 29. Research showed that homicide rates among youth aged 10–24 increased in many parts of the world between 1985 and 1994 (Legge 2008). The research literature provides a wealth of approaches, focusing on different levels of explanations: individual, situational, and social cultural and economic factors. Among these factors, prior research often emphasizes the importance of the socioeconomic status of the family and of youth. According to Legge (2008), the international labor data showed that working poverty among youth was especially prevalent in sub-Saharan Africa and South Asia.

Some criminologists contend that the effect of age on crime is invariant across crime categories and cultural conditions (Gottfredson and Hirschi 1990). In other words, the age effect on crime is everywhere and at all times the same because crime declines as age increases. Empirical studies on homicide rates in some Asian countries suggest that this thesis needs to be modified.

Johnson (2008), for example, described Japan’s postwar homicide decline and the vanishing young killers in particular. Japan’s homicide rate has dropped 70% and become one of the lowest homicide rates in the world. According to Johnson, the World Health Organization data showed that the Japan’s homicide rate was 0.6 per 100,000 population in 2002, a rate that was about half the mean and the median for other developed nations. Comparing with other Asian countries, Japan also had few homicides. According to Hasegawa and Hasegawa (2000), Japan’s postwar decline in homicide rate has made it an extraordinary country.

According to Johnson (2008), the most plausible explanation for Japan’s homicide drop was best described demographically. Johnson found that Japan was a striking exception to the pattern found in other nations where young males were most likely to kill. Data showed that, in contrast, the most murderous demographic in Japan was men in their 40s and 50s. In fact, Japan’s homicide rate in 2000 was higher among men in their 50s than among males aged 20–24. Further, young Japanese males, those born after 1960, committed fewer murders than previous cohorts of young males did. Data showed that young Japanese males now committed one-tenth as many homicides as their counterparts did in 1955. Thus, young Japanese males not only commit fewer murders than youth in other countries, they kill far less frequently than their predecessors did in previous decades.

it is not common to find countries where the propensity to kill peaks so late. However, Johnson (2008) found that South Korea resembled Japan. Johnson collected data from the reports by the Supreme Prosecutors Office of South Korea. According to the official data, in 2004, only 13.1% of suspects indicted for homicide in South Korea were aged 30 or under. Compared with the homicide rate in Japan, South Korea’s total homicide rate (excluding attempts) was a little higher at the rate of 1.67 per 100,000 in 2004. Taken together, the findings on homicide from Japan and South Korea ­suggest that how homicide rates are patterned can vary across countries, and the different ­patterns in Japan and South Korea are indeed noteworthy.

The Japanese decline is a notable trend, because homicide tends to increase during the industrialization process (LaFree and Drass 2002). However, very limited research has studied the homicide drop in Japan and especially the homicide drop among young Japanese. This body of research attempted to offer some explanations. For example, Roberts and LaFree (2004) studied the declining levels of economic stress in postwar Japan on homicide rates. Extant explanations also include social structure (Park 2006) and culture (Komiya 1999).

Hiraiwa-Hasegawa (2005) developed a risk-assessment explanation. They argue that young males became more risk averse in the postwar period. During this period time, income and educational level rose, while inequality and the average number of children per household declined due to the nation’s postwar economic achievements. Further, In Japan, homicide was a very risky way of resolving conflict. First, there is a high probability that would be assailants will themselves be injured by acts of self-defense. In Japan, weapon is used in most homicides, and about 60% of all instruments of death are knives. In contrast, guns are used in only about 5% of homicides. Second, the homicide clearance rate in Japan is extremely high, and it is very likely that a successful killer will be caught by the police. Data show that the Japanese police clear about 96% of all homicides reported to the police, while the chance of getting away with murder is about 1 in 3 for America, in 4 for France, and just 1 in 25 for Japan (Finch 2001). For the above reasons, it is argued that murder is relatively rare in Japan.

8 Population Diversity and Homicide

Population diversity can be described along two dimensions: heterogeneity and inequality (Avison and Loring 1986). Heterogeneity is based upon the distribution of a population among groups in terms of nominal parameters such as ethnicity, race, language, or religion. Inequality refers to status distribution in terms of a graduated parameter such wealth, income, or education (Blau 1977).

Cross-national studies have often reported a positive correlation between income inequality and homicide rates. This correlation suggests that in a country where income inequality is high, homicide rate is also high, and this correlation has been observed by studies employing different samples of nations, different measures of variables, and different control variables (Braithwaite and Braithwaite 1980; Kick and LaFree 1985; Messner 1980). However, the evidence concerning the effects of heterogeneity on homicide is less clear.

Using homicide data from the World Health Organization from 1967 to 1971, Avison and Loring (1986) examined the impact of income inequality and ethnic heterogeneity on homicide rates for a sample of 32 nations and regions. This sample included Japan, Hong Kong, Philippines, Thailand, and Taiwan. They found that the two dimensions of population diversity had significant main effects on cross-national homicide. Specifically, as heterogeneity and inequality increase, the homicide rate increases. In addition, increased ethnic heterogeneity exacerbated the impact of income inequality on homicide rates.

9 Democracy and Homicide

There were simultaneous increases in democratization and violent crime rates in many countries during the second half of the twentieth century. According to Potter et al. (1997), this rapid political transformation began in Southern Europe in the 1970s, spread to Latin America and parts of Asia in the 1980s, and then moved on to areas of sub-Saharan Africa, Eastern Europe, and the Soviet Union in the late 1980s and early 1990s.

LaFree and Tseloni (2006) studied whether the wave of democratization in the last half of the twentieth century could explain the global rise in violent crime rates. They collected annual time-­series data from WHO on homicide victimization rates per 100,000 population in 44 countries for varying years between 1950 and 2000. Asian countries such as Japan, Israel, Philippines, Singapore, and Thailand were included in the analysis.

In this study, the democracy measure was derived from evaluations of four characteristics of national governments. These four characteristics included the competitiveness of political participation, the openness of executive recruitment, the competitiveness of executive recruitment, and the constraints on the chief executive.

According to LaFree and Tseloni (2006), violent crime rates are curvilinear with the highest rates in countries that are transitioning between autocracy and democracy. In other words, homicide rates of full democracies on average may not be significantly different from countries with autocratic governments. Further, countries moving from autocratic to transitional democracies will experience a significant increase in homicide rates.

10 Modernization and Homicide

The modernization perspective asserts that crime booms are inevitable results of the rapid social changes experienced by nations that are in transition from traditional to modern forms of organization (Lafree and Drass 2002). According to this perspective, crime results when modern values and norms come into contact with and disrupt older, established systems of role allocation. During the transitional period, the emerging new roles are not fully institutionalized and integrated into society, making normative guidelines ambiguous, which in turn disrupt the traditional forms of social organization. Involved in this transitional process are a wide range of concepts in criminological theories, including social disorganization, anomie, breakdown, tension, and strain. These concepts help explain why modernization is linked to rising crime rates and other forms of deviance. It suggests that crime booms are most likely to occur in industrializing societies as they are shifting from agricultural to industrial and service economies (LaFree and Drass 2002).

According to LaFree and Drass (2002), the modernization perspective emphasizes crime booms which are defined as positive, rapid, and lasting changes in crime rates. To explain crime booms and test the effects of industrialization and modernization, LaFree and Drass used annual homicide victimization data of 34 nations from 1956 to 1998 and assembled an annual time-series dataset of WHO homicide victimization rates. The sample included three industrialized countries and regions from Asia, and they are Singapore, Hong Kong, and Japan. However, all the industrializing nations in the sample were from east European and Latin American countries. In support of modernizaitonalists, this comprehensive study found that 70% of industrializing nations had homicide booms.

In contrast, fewer than 21% of industrialized nations did. Thus, Lafree and Drass concluded that homicide trends were a more urgent contemporary problem in the industrializing countries than in the industrialized countries. However, the implications of this finding for Asian countries are limited because none of the industrializing countries in Asia were included in this analysis.

11 Social Stress, Support, and Homicide

Agnew (1992) points out that failure to achieve expectations may lead to such emotions as anger, resentment, rage, dissatisfaction, disappointment, and unhappiness, which are emotions customarily associated with strain in criminology. In his general strain theory, Agnew emphasized strain as the key concept for understanding and explaining crime and delinquency in general. Macro-level empirical research has considered violent crime to be major possible response to social stress. For example, in two comparative analyses of social stress in the 50 states of the USA, it is reported that stressful life events were positively correlated with all seven index crimes. Specifically, the higher the state stress level, the higher the crime rates for these states (Linsky and Straus 1986; Linsky et al. 1995).

Israel has been studied to explore the influences of social stress and social support on homicide and other violent behavior. Landau (1988) investigated at the aggregate level the relationship between the subjective perception of social stress and support and violence (i.e., homicide and robbery) in Israeli society. The analysis was based on monthly data for the years 1967–1979. Violent crime was positively related to most of the subjective stress indicators and negatively related to the subjective perception of national solidarity.

Landau (1997) reexamined the effects of social stress and social support on homicide for the years 1979–1993. According to Landau, replication over time was particularly important in the case of Israel, due to the dynamic and highly stressful characteristics of this society. The period of 1979–1993 was characterized by major crises and dramatic events in Israel’s security and economy. The most salient were the war in Lebanon, the Palestinian uprising, and the Gulf War. In addition, these years witnessed times of economic prosperity as well as economic hardship, recession, dramatic increases in the rate of inflation, labor disputes, prolonged strikes, terrorist attacks, and other acts of violence.

In the follow-up study by Landau (1997), the independent variables related to stress included the state of the country and government handling (regarding general situation, economic situation, security situation, and political situation) and personal situation (personal economic situation), coping, and mood. The independent variables related to national solidarity included solidarity between ethnic groups, solidarity between religious and secular groups, and readiness for economic sacrifices. This study found that economic stress had the most permanent and consistent effects over time on homicide, and the effects of social solidarity on homicide were stable over time. The consistent and permanent effect of ­economic stress on homicide in both studies is especially noteworthy. The findings are also in line with a wide range of studies connecting objective economic stress factors such as inflation and unemployment with homicide.

12 Suicide and Homicide

There are patterns about suicide and homicide in a society, and some social factors may be correlated with both suicide and homicide. In a study of 64 countries and regions, including China, Hong Kong, Israel, Japan, Korea Republic, Singapore, Macau, and other countries in the world, He et al. (2003) examined lethal violence (i.e., suicide and homicide) during 1989–1993. The predictors of this study included factors such as income inequality, economic development, unemployment rate, and divorce rate.

He et al. (2003) found that income inequality and economic development were not related to lethal violence but were related to the tendency of suicide over homicide. In particular, high inequality reduced the tendency to express frustration inwardly against oneself and increased the tendency to express frustration outwardly against other people. Further, the economic development increased the tendency to express one’s frustration inwardly against oneself and reduced the tendency of expressing one’s frustration against others. Finally, they found that divorce increased both suicide and homicide in the contemporary societies of the world.

Consistent with the findings of He et al. (2003), Japan’s homicide and suicide rates are noteworthy. As a country of one of the lowest homicide rates in the world, Japan has a high suicide rate. According to Johnson (2008), postwar developments have made Japan affluent without creating extreme social inequalities or the concentrations of poverty. Empirical research shows that homicide problems are usually correlated with social inequality, and thus homicide is not a serious problem in Japan. During the postwar period, the homicide rate had reduced significantly. In contrast, suicide rate is high in Japan. The combined rate of lethal violence (i.e., homicide and suicide) in Japan exceeds that for every other industrialized nation, and it is about twice the average for all industrialized countries (Johnson 2008). Particularly, males commit about three quarters of suicides in Japan, and since 1998 the nation’s suicide rate has risen substantially among several subpopulations, including young men (Miyazaki and Otani 2004). According to Johnson (2008), about 40 times more Japanese kill themselves than kill other people.

13 Homicide Clearance

Clearance rate is a measure of crimes solved by the police, and homicide clearance is usually relatively high among all the categories of violence. However, homicide clearance varies across country, and the Japan’s high homicide clearance rate is remarkable. In a comparative study, Roberts (2008) shows that homicide clearance rates in Japan and USA are stable between 1994 and 2004. However, Japanese homicide clearance rates were around 95% compared with roughly 60% in the United States.

There are a number of reasons for the extremely high homicide clearance rate in Japan. Studies have focused on the high level of citizen cooperation and the good police–citizen relationship in Japan which are essential parts of homicide investigation (Bayley 1991; Fujimoto 1994; Rake 1987; Litwin 2004; Riedel and Jarvis 1998). In addition to these studies, Roberts (2008) has focused on the characteristics of homicides in Japan and the categorization of homicide in official statistics as an explanation for the much higher homicide clearance rate in Japan than in other countries such as the United States.

Using official summary statistics from 2000 to 2004, Roberts (2008) found that homicides in Japan contained a higher proportion of easy-to-clear cases, including those with nonfirearm weapons, family member offenders, and child victims. First, homicide clearance studies have found that cases involving a firearm weapon are less likely to clear than those involving a knife or strangulation, because of the lack of close contact with the victim and less physical evidence (e.g., offender’s hair, blood, and fingerprints) (Regoeczi et al. 2000; Litwin 2004; Puckett and Lundman 2003; Addington 2006). In Japan, the general public is banned from owning firearms. Between 2000 and 2004, only 3.4% of homicides in Japan were committed with firearms, and most homicides were committed with knives, blunt objects, or by strangulation. In contrast, 65.7% of the homicides in the United States were committed with firearms. Second, homicides committed by known offenders, especially family members, have much greater chance of clearance than homicides by strangers (Riedel and Rinehart 1996; Roberts 2007). Offenders in family-related homicides are more likely to confess than offenders unknown to the victims. Also, family-related homicides tend to have more potential eyewitnesses. Third, homicide incidents with child victims have usually been found to have greater chance of clearance due to the restricted routine activities of children (Regoeczi et al. 2000; Puckett and Lundman 2003; Addington 2006; Riedel and Rinehart 1996).

Roberts (2008) also found that categorization of homicides in Japanese official statistics included attempted homicides and excluded robbery-related homicides, which could also explain the differences between the clearance rate for homicide in Japan and that in other countries. In Japan, robbery-related homicide cases were less likely than other homicides to be cleared. Between 2000 and 2004, non-robbery-related homicide clearance rates averaged 94.7%, whereas robbery-related homicide clearance rates averaged 84.9%. In recent years, about half of the Japanese cases categorized as “homicide” were actually attempted homicides. However, the clearance rates for the two types of offenses in Japan are not very different: 95.8% for attempted homicides and 93.4% for completed homicides. In sum, though categorization of homicides in Japan may be different when comparing with other countries, the overall clearance in Japan is consistently high.

14 Conclusion

There are a number of international homicide data sources, and three data sources including Interpol, UN-CTS, and WHO are often used by scholars to describe international homicide trends. These data sources sometime provide different estimates, because there are definitional differences among these data sources. In addition, countries vary in their legal definitions and reporting practices to the international data sources. Using these data sources, empirical research has explored homicide trends in Asia and offered a number of explanations for the homicide trends. These studies are usually at the macro-level, and the factors that have been studied include firearms, capital punishment, population characteristics (such as youth population and population diversity), and social characteristics (such as democracy, modernization, and social stress). Homicide clearance is also analysis with a comparative approach.

Though empirical research has offered a broad range of explanations for homicide trends, not all the Asian countries have been extensively analyzed. Some countries and regions (such as Japan, South Korea, Singapore, Taiwan, Israel, Hong Kong) are more studied than other countries in the comparative research. Future research should address this limitation by improving current cross-national data sources on homicide and including more Asian countries in the comparative analyses.